Cellulose Structure and Biosynthesis: What is in Store for

the 21st Century?

R. MALCOLM BROWN, JR.

Molecular Genetics and Microbiology, University of Texas at Austin, Austin, Texas 78712

Received 12 February 2003; accepted 13 June 2003

ABSTRACT: This article briefly summarizes historical developments in fundamental

research related to the structure and biosynthesis of cellulose. Major advances concern-
ing the structure of cellulose include the discovery of a new suballomorph of cellulose I,
the lattice imaging of glucan chains showing no fringe micelle structure, parallel chain
orientation in cellulose I, and the discovery of nematic ordered cellulose. Major ad-
vances in biosynthesis include the discovery of the terminal synthesizing complex, the
isolation and purification of cellulose synthase, the in vitro synthesis of cellulose I, and
synthetic cellulose assembly. This article focuses on recent advances in molecular
biology with cellulose, including the cloning and sequencing of cellulose synthase genes
from bacteria, cyanobacteria, and vascular plants; proof of the terminal synthesizing
complex as the site of the catalytic subunit of cellulose synthase; cellulose and callose
synthase expression during growth and development; and phylogenetic aspects of
cellulose synthase evolution. This article concludes with thoughts about future uses for
the accumulating genetic information on cellulose biosynthesis for textiles and forest
products and discusses possibilities of new global resources for cellulose production.
© 2003 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 42: 487– 495, 2004
Keywords: cellulose; structure; biosynthesis; molecular genetics; new cellulose crops;
biomaterials; biopolymers

INTRODUCTION markable that more attention has not been de-

I am pleased to dedicate this review to my col-
league and good friend, Professor Otto Vogl, who
has made seminal contributions to the field of
polymer chemistry (Fig. 1). I am also honored to
be invited to speak in Vienna on a subject very
dear to my heart: cellulose (Fig. 2). This biopoly-
mer is the most abundant macromolecule on
earth, with more than 1011 tons estimated to be
synthesized each year on our planet.1 Given the
vast quantity of this product and the importance
of this material to humankind, it is indeed re-
Correspondence to: R. M. Brown,
rmbrown@mail.utexas.edu)
Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 42, 487– 495 (2004)
© 2003 Wiley Periodicals, Inc.
voted to the scientific exploration of cellulose, es-
pecially with respect to its structure and biosyn-
thesis. Although this was one of the earliest
molecules to be subjected to X-ray analysis,2 and
considerable efforts have been directed toward its
structure during the past century, only now are
we beginning to unlock the secrets of this fasci-
nating biopolymer.
I will present a brief overview of cellulose, fin-
ishing with what is in store for us as we move into
the 21st century. Indeed, in the era of synthetic
polymers, cellulose continues to be a dominant
force in many vital industries, including the pulp
and paper, lumber, and textile industries. Will it
Jr. (E-mail: maintain this dominance? What factors will come
into play as we develop post-genomic-era engi-
neering methods for new cellulose crops? I will
487

488 BROWN
Figure 1. Otto Vogl visiting the author’s home in
Austin, Texas, in May 2001.
cover these and other relevant questions in this
brief review. The references in this article will
direct readers to the most recent literature in the
field for further study. The reader is also encour-
aged to visit my website for additional references
and information.3
DIVERSITY OF CELLULOSE: WHERE IS IT
FOUND?
Cellulose is synthesized by a great diversity of
living organisms. In fact, we usually associate our
resources as trees and cotton plants, and right-
fully so, because these are the major sources of
the industrial production and processing of cellu-
lose. However, cellulose is synthesized by bacteria
and prokaryotes as well (e.g., Acetobacter, Rhizo-
bium, and Agrobacterium). Even some pathogenic
bacteria have been found to synthesize cellulose.4
One of the most interesting recent discoveries is
that the most ancient forms of life on earth, rep-
resented by the cyanobacteria, also synthesize
cellulose5 (Fig. 3). Of course, eukaryotic organ-
isms produce cellulose, and certain fungi, amoe-
10990518, 2004, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pola.10877 by INASP (Tanzania), Wiley Online Library on [14/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
bae, cellular slime molds, and green algae have
representatives that produce perfectly good cellu-
lose.6 In fact, one of nature’s most perfect crystal-
line forms of cellulose comes from a green alga,
Valonia ventricosa; more than 1200 glucan chains
produce very highly ordered crystalline microfi-
brils with a degree of polymerization (dp) of more
than 25,000.1,7 Cellulose is also synthesized by
freshwater and marine algae, including land
plants such as mosses, ferns, angiosperms, and
gymnosperms.1 Cellulose is even made by some
animals, the tunicates.8 Why is there all this di-
versity? This is highly suggestive of an ancient
evolutionary process that has culminated in the
production of diverse forms of cellulose among a
great variety of organisms on earth. Therefore, it
is logical to conclude that on a functional basis,
cellulose is essential for many cellular functions.
DIVERSITY OF CELLULOSE BIOSYNTHESIS:
HOW LONG HAS IT BEEN AROUND?
Cellulose biosynthesis must be an ancient pro-
cess. As we learn more about the phylogeny of
living organisms and the origin of life on earth,
we cannot come away unimpressed by the ex-
treme age of life on our planet. It has been around
for more than 3.5 billion of the 5– 6 billion years
that the earth has existed.9 The earliest known
life forms may have been photosynthetic or che-
mosynthetic prokaryotic organisms. The stroma-
tolites found in Canada and Australia9 are com-
posed of algal mats, mostly cyanobacteria. Cya-
nobacteria have remained relatively unchanged
for billions of years. Recently, David Nobles in my
laboratory found that cellulose biosynthesis is
widespread among the various cyanophycean
genera,5 and this implies that cellulose has been
around a long time and has probably dramatically
influenced the evolution of life. Did the first form
of life produce cellulose? We do not have an an-
swer to this provocative question, but as we learn
more through gene sequencing and genome anal-
ysis, it may be possible to reconstruct life’s tree
back to an even earlier time than we know from
fossil evidence, a time when life may have just
begun. I think that cellulose may have had an
important role in the successful formation of life
on earth, particularly in sustaining life forms in
earth’s harsh primitive atmosphere. I have dis-
cussed this in a preliminary way recently, and I
believe that cellulose may have offered protection
from dangerous ultraviolet radiation when the
 10990518, 2004, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pola.10877 by INASP (Tanzania), Wiley Online Library on [14/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
CELLULOSE STRUCTURE AND BIOSYNTHESIS 489

Figure 2. The top image is the structural formula of cellulose. The arrows point to the
basic repeat unit, which is a cellobiose molecule. The molecule has a twofold screw–axis
symmetry. The bottom image is a schematic diagram; the lines show the individual
glucan chain polymers that constitute the crystalline cellulose I microfibril.

earth still had a reducing atmosphere devoid of cose units per chain (dp ⫽ 8000). Secondary wall
oxygen.5 Cellular evolution is a hot new field of cellulose has a higher dp, up to 15,000.
science that will certainly benefit from an under- Native cellulose is found in two crystalline
standing of the biosynthetic pathways and their forms, cellulose I and cellulose II. By far, most
roles in survival. That is not the major topic of
this review, but it has sufficient merit that I
would urge the reader to bookmark this section
and to follow up on subsequent publications as
they appear.

STRUCTURE OF NATIVE CELLULOSE

Native cellulose is defined as that cellulose made

by living organisms. Of course, it can be altered
by strong alkali treatments to produce other crys-
talline forms. Cellulose is chemically composed
only of glucose monomers (Fig. 2). The linkage is
␤-1,4. Starch is composed only of glucose, but the
linkage is ␣-1,4. The number of glucose mono-
Figure 3. Cellulose microfibrils from a cyanophycean
meric units required to produce an insoluble prod-
alga, Nostoc muscorum, negatively stained with uranyl
uct is about 8. Above that, the glucan chains have acetate. The opaque electron spheres are 10-nm gold
a greater affinity for one another than they do for particles bound to the enzyme CBH-I, which specifi-
the aqueous solvent. A typical number of glucose cally binds to the crystalline surface of cellulose I. Re-
units in native cellulose depends on the source, printed with permission from D. Nobles, D. Romanov-
such as primary or secondary cell walls. Primary icz, and R. M. Brown, Jr. 2001. Plant Physiology 127,
cell wall cellulose polymers have about 8000 glu- 529 –542. © 2001 American Society of Plant Biologists.
 10990518, 2004, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pola.10877 by INASP (Tanzania), Wiley Online Library on [14/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
490 BROWN

possible to image single glucan chains in this

cellulose because they are surrounded by the ura-
nyl acetate negative stain (Fig. 5).
The history of the structure of cellulose is long
and contorted. One of the earlier favored models
for native cellulose was a microfibril with a crys-
talline core surrounded by a fringe micelle.14 We
know from recent lattice image studies using
HRTEM15 and atomic force microscopy (AFM)16
that the crystalline arrangement of the glucan
chains is constant and subtends the entire micro-
fibril structure. There are no fringe micelles. An
image of the lattice structure of glucan chains in
a single microfibril of Boergesenia is shown in
Figure 4. The structure of cellulose II is not so
Figure 4. Lattice image of a microfibril from a green straightforward. The glucan chains that we know
alga, Boergesenia forbsei. The signal-to-noise ratio is are antiparallel, but are they long chains, or are
not high for the imaging of the single glucan chains, but
they folded chains? One relatively recent study17
they can be detected. The 0.53-nm glucan chain spac-
ings are revealed in this image. Reprinted with permis- showed that in a mutant of Acetobacter, cellulose
sion from S. Kuga and R. M. Brown, Jr. 1987. J Elec- II was arranged in the form of folded chains. The
tron Mic Tech 6, 349 –356. © 1987 Oxford Press. situation may be very different in a nonnative
form of cellulose II known as rayon, which is
produced from the reprecipitation of alkali-solu-
native cellulose exists as cellulose I. The glucan bilized cellulose. Rayon is known to have a rela-
chains are oriented parallel in cellulose I,10 but in tively low dp (ca. 300 –500), and the nanostruc-
cellulose II, the chains are antiparallel. The ther-
modynamically most stable allomorph is cellulose
II, which has an additional hydrogen bond per
glucose residue. Native cellulose II is rare and is
found only in several algae as well as some bac-
teria.11 Cellulose I allomorphs consist of distinct
numbers of parallel glucan chains arranged to
form the nanostructure known as a microfibril.
The number and arrangement of the glucan
chains are under the genetic control of the en-
zyme complex that synthesizes the cellulose. Mi-
crofibrils come in many shapes and sizes, ranging
from thin membrane-like structures, as found in
Erythrocladia,12 to giant square structures with
more than 1200 glucan chains, as found in Valo-
nia,1 or large rectangular microfibrils with many
hundreds of glucan chains, as found in Boergese-
nia (Fig. 4). Furthermore, the reducing ends of
glucan chains in a microfibril are found in the tips
of microfibrils away from the site of synthesis on
the cell.10
I might also add that a noncrystalline form of
cellulose has been recently described. It is called
Figure 5. Nematic ordered cellulose imaged by
nematic ordered cellulose.13 This cellulose is HRTEM after negative staining with uranyl acetate.
highly ordered but noncrystalline and is produced The individual glucan chains are ordered but are not
from solution in lithium dimethylacetamide by tightly associated to form a crystalline phase. Re-
the slow introduction of water vapor (and con- printed with permission from T. Kondo, E. Togawa, and
trolled hydrogen bonding). With high-resolution R. M. Brown, Jr. 2001. Biomacromolecules 2, 1324 –
transmission electron microscopy (HRTEM), it is 1330. © 2001 American Chemical Society.

Figure 6. AFM height image of a bacterial cellulose
subunit surface acquired by the contact mode in air,
which may correspond to the intermolecular and in-
tramolecular periodicity for the (100) crystal face of the
triclinic cellulose I␣ form of cellulose (photograph and
preparation by R. M. Brown and T. Kondo).
tures are not imaged as long microfibrils but
rather as short fibrils with tapering ends.18
Native crystalline cellulose I also has two dif-
ferent suballomorphs, cellulose I␣ and cellulose
I␤.19 Of these, cellulose I ␤ is the more thermody-
namically stable, and it is rarely synthesized in
nature in a pure form, with the exception of tuni-
cates. Cellulose I␣ exists as a single-chain tri-
clinic unit cell, whereas cellulose I␤ has a two-
chain monoclinic unit cell. These structures have
been imaged with HRTEM16 and AFM (Fig. 6).
FUNCTION OF NATIVE CELLULOSE
Although it may appear obvious that the struc-
ture of cellulose is for the protection of a cell, I
would like to briefly elaborate to provide new
insight into this concept. Certainly, there can be
no argument that a coating or covering of cellu-
lose to produce a cell wall is of great importance in
protecting the delicate protoplasm from the envi-
ronment. Equally important are the site of depo-
sition and the patterns of the arrangement of
microfibrils in the cell wall. All plant cells enlarge
throughout their cell cycles. This is called growth.
Plant cell growth can be random, isodiametric, or
directional, depending on the arrangement of the
10990518, 2004, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pola.10877 by INASP (Tanzania), Wiley Online Library on [14/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
CELLULOSE STRUCTURE AND BIOSYNTHESIS 491
cellulose. For example, in tip growth in apical
cells, root hairs, or pollen tubes, the polymer ori-
entation is random. If a cell is genetically pro-
grammed to enlarge by producing a tube, the ar-
rangement of cellulose microfibrils is critical to
the growth leading to the elongation of the tube.
The microfibrils are first deposited in the axis
transverse to the cell’s long axis; however, as the
tube grows or enlarges by the forces of the turgor
pressure from within, the transverse microfibrils
shift toward the longitudinal axis of the cell. More
frequently, the cellulose microfibrils become de-
posited in a helical fashion (described later with
respect to an ordinary child’s toy). When roots,
stems, and hypocotyls elongate and grow, this is
the result of the sums of individual elongations of
the individual cells. Cellulose deposition often
controls cell elongation by occurring in a specific
pattern on the cell surface. The control of this
deposition pattern is linked to the cell’s cytoskel-
etal system, but for now, consider as a model for
this concept, if you will, a Slinky. This toy is often
used to jump down stairs. When a Slinky is
pulled, it is elongated, but its diameter remains
constant. The metal spiral coil merely changes its
transverse orientation to a more parallel orienta-
tion. The diameter of the Slinky remains un-
changed. In the real world of the plant cell, the
turgor pressure is the driving force for cell en-
largement or growth. Growth is defined as an
irreversible increase in the cell volume. If one
could place a balloon inside a Slinky and then
blow it up, the balloon would only be able to
expand to form a cylinder. In the real world, di-
rected growth is cell elongation, as demonstrated
by the Slinky model. A cellulose microfibril by
itself cannot be extended. The nanostructure is a
rigid rod; however, microfibrils can slide past
neighboring microfibrils if a noncellulose grease
is present. In plant cell walls, other polymers
such as xyloglucans and rhamnoglacturonans
form this grease. Cell wall expansion and growth
is a very complex subject, and I will not describe
it further, except to point out the central role for
cellulose as the skeleton for maintaining and con-
trolling the growth processes.
BIOSYNTHESIS OF NATIVE CELLULOSE:
NATURE’S NANOMACHINE PAR
EXCELLENCE
At first thought, it seems rather simple that a
homopolymer with a specific linkage would re-
 10990518, 2004, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pola.10877 by INASP (Tanzania), Wiley Online Library on [14/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
492 BROWN

quire only a simple enzyme to polymerize glucose

into ␤-1,4-linked chains. The case is not so simple,
for as we have learned, glucan chains are highly
ordered into nanostructures called microfibrils.
How does the cell’s machinery not only accom-
plish the simultaneous polymerization of multiple
glucan chains but also arrange these into meta-
stable crystalline nanostructures known as mi-
crofibrils?
Before the 1950s, no one had any good ideas
about the biosynthetic machinery for cellulose;
however, as biological electron microscopy ap-
proaches became more widespread, attention
turned toward the role of the plasma membrane
in cellulose biosynthesis. After all, the cellulose
Figure 7. The top image is the E-fracture face of a
microfibrils were observed on the outside of the
plasma membrane of Oocystis apiculata showing a lin-
cell, yet they were in close contact with the cell’s
ear cellulose synthesizing TC. There are three rows of
contents; therefore, it seemed logical that the cel- subunits that comprise this complex. The TC moves in
lular biosynthetic machinery also would be in- the plane of the plasma membrane as the cellulose is
volved in the extracellular materials comprising generated and attaches to the inner cell wall layer. This
the cell wall. In 1958, Roelofsen20 proposed that alga was the first organism for which TCs were de-
the cellulose microfibril originated from an en- scribed. No fixation was used. Reprinted with permis-
zyme complex and that it was assembled by tip sion from R. M. Brown, Jr. and D. Montezinos 1976.
growth; however, no direct proof for such a struc- Proc Nat Acad Sci 73, 143–147. © 1976 National Acad-
ture had been forthcoming. In 1976, my graduate emy of Science. The bottom image is the P-fracture face
student David Montezinos and I discovered an of a plasma membrane of a Zea mays root showing
several rosette TCs. Unlike the linear TC, the subunits
enzyme complex associated with the ends of grow-
are hexagonally arranged. Rosette TCs are common to
ing microfibrils in a green alga, Oocystis apicu-
all vascular plants, including some algae (Chara, Ni-
lata.21 The development of an electron microscopy tella, Spyrogyra, and Micrasterias; courtesy of Susette
preparative method known as freeze fracture Mueller)
made this discovery possible. Living cells are rap-
idly frozen in Freon and then are placed in a high
ing from algae to tunicates to bacteria and
vacuum. They are fractured with a microtome
plants.22 These enzyme complexes are termed ter-
knife, and the fracture surface is etched by vac-
minal complexes (TCs). In the case of Oocystis,
uum sublimation. A platinum– carbon film evap-
TCs are called linear because they have three
orates on the etched surface. A carbon backing
rows of particles. These represent not only the
film is applied, and the sample is brought to room
catalytic subunits involved in cellulose assembly
temperature and atmospheric pressure. The rep-
but also ancillary proteins that help to position
lica is removed from the sample, cleaned, and
the glucan chains to achieve precise ordering in
examined in a transmission electron microscope.
the metastable crystalline state. In vascular
The unique principle of freeze fracture is that the
plants, which assemble cellulose in trees and cot-
hydrophobic interiors of the cell membranes often
ton, for example, TCs are termed rosette because
are fractured, revealing the transmembrane pro-
unlike Oocystis, they have six hexagonally ar-
teins. Thus, images of ordered structures shown
ranged subunits in the plasma membrane (Figs. 7
to be directly associated with impressions of cel-
and 8). The rosette TC in vascular plants was first
lulose microfibrils offer tangible evidence that a
described by my graduate student Susette Muel-
multienzyme complex in the plasma membrane is
ler in 1980.23
responsible for cellulose biosynthesis. In the spe-
cific instance with Oocystis, we found that the
ordered enzyme complexes consisted of three lin- GENES INVOLVED IN CELLULOSE
ear rows of particle subunits (Fig. 7, top). BIOSYNTHESIS
Since 1976, freeze fracture has revealed a mul-
titude of enzyme complexes associated with cellu- With tremendous progress in molecular genetics
lose microfibrils from a variety of organisms rang- in the 1980s, the stage was set for gene sequenc-

Figure 8. Hypothetical model of a rosette TC show-
ing the possible origin of glucan chains. This TC pro-
duces a crystalline microfibril with approximately 36
glucan chains. Each subunit may synthesize a 6-glu-
can-chain sheet held together by van der Waals forces.
The sheets then stack by hydrogen bonding to form the
three-dimensional microfibril.34 –36 Reprinted with per-
mission from R. M. Brown, Jr. and I. M. Saxena 2000.
Plant Physiology and Biochemistry 38, 57– 67. © 2000
Elsevier.
ing on a much larger scale, and in 1990, my col-
leagues Inder Saxena and Fong Chyr Lin and I
first isolated and sequenced a gene for cellulose
biosynthesis from Acetobacter xylinum.24 During
the past 13 years, genes for cellulose biosynthesis
have been characterized from a great variety of
organisms, including vascular plants.25 I will not
go into details here, except to note that the se-
quence of genes for cellulose biosynthesis has led
to notable comparisons of highly conserved se-
quences among many organisms with the discov-
ery by hydrophobic cluster analysis26 that the
aspartate residue is involved in the catalytic re-
action and that two domains and three aspartates
are required for cellulose biosynthesis. In addi-
tion, a QXXRW motif regulates the catalysis. Mu-
tations of any part of this conserved sequence lead
to abnormal plant growth and development and
altered cellulose assembly.27 Notably, the RSW-1
radial root swelling mutant of Arabidopsis
greatly reduces crystalline cellulose and produces
instead noncrystalline cellulose.28
10990518, 2004, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pola.10877 by INASP (Tanzania), Wiley Online Library on [14/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
CELLULOSE STRUCTURE AND BIOSYNTHESIS 493
An excellent example of a spin-off from the
genetic data comes from the work of one of my
former graduate students, Walairat Laosinchai,
who cloned a cotton cellulose synthase gene, ex-
pressed it in E. coli, and then isolated the genet-
ically engineered cotton cellulose synthase from
E. coli and produced antibodies against it.29 Us-
ing a difficult exploitation of freeze fracture
known as fracture labeling, Kimura and Itoh,30
along with members of my laboratory, were suc-
cessful in immunolabeling the rosette TCs in a
vascular plant. This study provided the best di-
rect evidence to date that the rosette TC contains
the catalytic subunit for cellulose synthase.
One of the most exciting aspects of cellulose
synthesis research lies in the genetic revolution
and what it can and is telling us about cellulose.
As I mentioned previously, one of my graduate
students, David Nobles identified cellulose syn-
thesis in the cyanobacteria, the oldest living ex-
amples of ancient life on earth. The genetic se-
quences also tell us something else. They suggest
but do not prove that cellulose synthase genes
were introduced into a primitive eukaryotic cell
from the cyanobacteria by lateral gene transfer.
This is a most interesting point, because if this is
really true, we then have an understanding of
how land plants evolved, including photosyn-
thetic eukaryotes! Obviously, more supporting ev-
idence will be needed to confirm this concept, but
the critical and key points already have been elu-
cidated. Cellulose synthase genes can be injected
into organisms and in this mode can dramatically
alter successful survival. The evolutionary his-
tory of the origin of life on earth will, in my
opinion, be elucidated by an understanding of
cellulose biosynthesis. I realize that this is a far-
reaching statement, but I believe that it is one
that will be shown to be of immense importance in
understanding not only cellular evolution but also
the evolution of many eukaryotic life forms, in-
cluding early animals. It is significant that the
recent genome sequence of the tunicate Ciona31
shows the possibility of the lateral gene transfer
of cellulose synthase into this animal. This sug-
gests that if this had not happened, the tunicate
would never have evolved or developed and we
would not have tunicates with cellulose! I expect
great advances to be made in the next few years
as we sequence more genomes and better under-
stand the role of cellulose biosynthesis from these
studies.

494 BROWN
GENETIC ENGINEERING AND
NANOENGINEERING AND WHAT THEY
CAN DO FOR THE FUTURE: WHAT TO
EXPECT
I conclude this brief review with what amounts to
speculations on the roles of genetic engineering
and nanoengineering and cellulose. What can
these great tools do for those who study and use
cellulose? They can provide a vast cadre of new
discoveries! In simplest terms, if we can geneti-
cally engineer cellulose biosynthesis into a photo-
synthetic, halophilic, nitrogen-fixing system, it
will be possible to form a new global resource for
cellulose. Imagine using hypersaline regions of
the planet in which practically nothing grows to
produce perfectly good cellulose! Some examples
that come to mind are the Dead Sea and the
Salton Sea. In addition, shallow basins of seawa-
ter can be constructed in desert areas with ex-
tended periods of sunlight; this would improve
the prospect for adapting unwanted sites for pro-
ductive cellulose biosynthesis. The scale-up of
this prospect would suggest that these new re-
sources could even supplant existing resources,
including trees and cotton crops. If this ever hap-
pened, the forests of the future could be allowed
maximum diversity, and this could help to pre-
vent the precious loss of species that is presently
occurring at an alarming rate through deforesta-
tion and tree harvesting, as well as the clearing of
land for ranching and small farming. I do realize
that such a proposal is not a utopia for civilized
life on earth, but it is one that could vastly enrich
our lives.
In addition to the aforementioned proposal, I
suggest that also we strive to improve cotton and
forest trees through molecular genetics. There is
so much that could be done with the modern ap-
proaches of genetic engineering not only to make
more cellulose for the crop but also to improve the
quality of cellulose. For example, increasing the
dp and being able to control the number of glucan
chains contributing to a microfibril could have a
dramatic impact on the strength of common ma-
terials such as paper, wood, and textiles. Chang-
ing the crystalline parameters of the cellulose
could offer new ways of dyeing and inking cellu-
lose products. I would also like to see the design of
artificial, functional cellulase synthase complexes
for the production of specialty cellulose and cellu-
lose products. I can envision factories in which
cellulose can be produced without the necessity of
living organisms, much like the refining of petro-
10990518, 2004, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pola.10877 by INASP (Tanzania), Wiley Online Library on [14/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
leum products, and used to produce plastics and
synthetic polymers. Such cellulose refineries may
raise the standards of product quality to such a
degree that we would not have to rely on synthetic
polymers for many of our needs. Maintaining a
natural product with biodegradable properties is
also of great benefit when we consider future di-
rections.
Finally, I would like to see nanoscale technol-
ogy advancements in the production of novel cel-
luloses. I can envision the development of large-
scale nanomachines that have the ability to poly-
merize glucose and assemble the polymer chains
directly into microfibrils of specific sizes, shapes,
and molecular weights. A good example is the
recent tremendous progress in dip pen nano-
lithography.32 There is no reason not to use com-
puter chip assembly techniques to handle enzyme
assemblies on a large scale, and then the products
of these reactions can be put to good use. Mi-
crofluidics33 is already well underway and could
handle many aspects of this vision.
Many visions of the future have been intro-
duced in this review, and this was my intent;
however, I would like to think about the future of
polymer science in the context of natural polymer
systems. We still have so much to learn from
nature! I am very much aware of the great ad-
vances that have been made over the years in
polymer science, but I also think that a return to
basics in trying to understand the truly complex
and integrated systems of polymer biosynthesis
will greatly help us in the design of future poly-
mer-generating approaches, whether native or
synthetic. Thus, I hope this brief discussion will
shed some light on a long overdue consideration of
the significance of understanding the fundamen-
tals of natural biopolymer synthesis as we move
forward into the 21st century.
The author thanks the Welch Foundation (F1217), the
U.S. Department of Energy (DE-FG03-94ER20145),
and the Johnson & Johnson Centennial Chair Funds
(University of Texas at Austin) for their past and con-
tinued support of his research on cellulose and atomic
and molecular imaging.
REFERENCES AND NOTES
1. Brown, R. M., Jr. Pure Appl Chem 1996, 10, 1345.
2. Nishikawa, S.; Ono, S. Proc Tokyo Math Phys Soc
(Tokyo) 1913, 7, 131.
3. For many additional aspects of cellulose, I direct
the reader to our website, which has relevant pa-

pers, research in progress, movies, links, and many
other things related to cellulose. The revised Cel-
lulose Electronic Network may be particularly use-
ful. Please visit http://www.botany.utexas.edu.
4. Zogai, X.; Nimtz, N.; Rodhe, M.; Bokranzl, W.;
Romling, U. Mol Microbiol 2001, 39, 1452.
5. Nobles, D.; Romanovicz, D.; Brown, R. M., Jr. Plant
Physiol 2001, 127, 529.
6. Brown, R. M., Jr. In Experimental Phycology 1:
Cell Walls and Surfaces, Reproduction, Photosyn-
thesis; Wiessner, G.; Robinson, D. G.; Starr, R. C.,
Eds. Springer-Verlag: Berlin, 1990; p 20.
7. Shibazaki, H.; Kuga, S.; Onabe, F.; Brown, R. M.,
Jr. Polymer 1995, 36, 4971.
8. Kimura, S.; Itoh, T. Protoplasma 1995, 186, 24.
9. Schopf, J. W.; Walter, M. R. In The Biology of
Cyanobacteria; Carr, N.; Whitton, B., Eds.; Black-
well: Boston, 1982.
10. Kuga, S.; Brown, R. M., Jr. Carbohydr Res 1988,
180, 345.
11. Canale-Parola, E. Bacteriol Rev 1970, 34, 82.
12. Okuda, K.; Tsekos, I.; Brown, R. M., Jr. Proto-
plasma 1994, 180, 49.
13. Kondo, T.; Togawa, E.; Brown, R. M., Jr. Biomac-
romolecules 2001, 2, 1324.
14. Frey-Wyssling, A. Science 1954, 119, 80.
15. Kuga, S.; Brown, R. M., Jr. J Electron Microsc Tech
1987, 6, 349.
16. Baker, A. A.; Helbert, W.; Sugiyama, J.; Miles,
M. J. Biophys J 2000, 79, 1139.
17. Kuga, S.; Takagi, S.; Brown, R. M., Jr. Polymer
1993, 34, 3293.
18. Okuda, K.; Li, L.; Kudlicka, K.; Kuga, S.; Brown,
R. M., Jr. Plant Physiol 1993, 101, 1131.
19. Atalla, R.; Vander Hart, D. L. Science 1984, 223,
283.
10990518, 2004, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pola.10877 by INASP (Tanzania), Wiley Online Library on [14/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
CELLULOSE STRUCTURE AND BIOSYNTHESIS 495
20. Roelofsen, P. A. Acta Bot Neerl 1958, 7, 77.
21. Brown, R. M., Jr.; Montezinos, D. Proc Natl Acad
Sci USA 1976, 73, 143.
22. Tsekos, I. J Phycol 2000, 35, 635.
23. Mueller, S. C.; Brown, R. M., Jr. J Cell Biol 1980,
84, 315.
24. Saxena, I. M.; Lin, F. C.; Brown, R. M., Jr. Plant
Mol Biol 1990, 15, 673.
25. Please visit http://cellwall.stanford.edu.
26. Saxena, I. M.; Brown, R. M., Jr.; Fevre, M.; Ger-
emia, R.; Henrissat, B. J Bacteriol 1995, 177, 1419.
27. Saxena, I. M.; Kudlicka, K.; Okuda, K.; Brown,
R. M., Jr. J Bacteriol 1994, 176, 5735.
28. Arioli, T.; Peng, L.; Betzner, A. S.; Burn, J.; Wittke,
W.; Herth, W.; Camilleri, C.; Höfte, H.; Plazinski,
J.; Birch, R.; Cork, A.; Glover, J.; Redmond, J.;
Williamson, R. E. Science 1998, 279, 717.
29. Laosinchai, W. Ph.D. Dissertation, University of
Texas at Austin, 2002.
30. Kimura, S.; Laosinchai, W.; Itoh, T.; Cui, X.;
Linder, R.; Brown, R. M., Jr. Plant Cell 1999, 11,
2075.
31. Dehal, P.; et al. Science 2002, 298, 2157.
32. Lee, K.-B.; Park, S.-J.; Mirkin, C. A.; Smith, J. C.;
Mrksich, M. Science 2002, 295, 1702.
33. Unger, M. A.; Chou, H.-P.; Thorsen, T.; Scherer, A.;
Quake, S. R. Science 2000, 288, 113.
34. For proof of the two-step crystallization model for
cellulose biosynthesis and an energy analysis prov-
ing the two-step crystallization model, see refs. 35
and 36, respectively.
35. Cousins, S. K.; Brown, R. M., Jr. Polymer 1997, 38,
897.
36. Cousins, S. K.; Brown, R. M., Jr. Polymer 1995, 36,
3885.